EDA on Ultra Marathon Dataset (1798 - 2019)

---

Table of Contents

1. Executive Summary
2. Project Goals
3. Data Sources
4. Steps Performed
5. Key Insights
6. Visualizations and Results
7. Quantitative Highlights
8. Next Steps

---

Executive Summary

This project analyzes ultra-marathon race data from 2019 to uncover trends in participant demographics, performance, and event characteristics. Insights focus on race distances, athlete performance, and gender-specific trends based on USA races in 2019.

---

Project Goals

• Clean and preprocess a large historical dataset of ultra-marathon events.
• Explore gender differences in participation and performance.
• Identify top-performing athletes and races based on speed and completion time.
• Analyze seasonal trends and how event distances influence athlete outcomes.
• Provide actionable insights to organizers, participants, and enthusiasts.

---

Data Sources

• Dataset size: 7,461,195 records with 13 features.
• Focused on USA races from 2019, filtering to standardized distances: 50km, 50mi, 100km, and 100mi.

---

Steps Performed

1. Data Cleaning:

   • Filtered for USA races and selected standard distances.
   • Addressed missing values and corrected data types.

2. Data Transformation:

   • Extracted useful fields like athlete age and country.
   • Converted performance times into numeric formats.

3. Visualization and Statistical Analysis:

   • Explored demographic trends, gender comparisons, and seasonal participation.
   • Analyzed top-performing events and athletes.

---

Key Insights

1. Demographics:

   • Male athletes dominate participation, especially in longer races.
   • Average participant age: 41 years.

2. Performance Highlights:

   • Fastest 50km finish time: 2.83 hours.
   • Best-performing event: Caumsett Park 50K Championships.

3. Race Trends:

   • 50km and 50mi races are the most popular.
   • Spring and Summer seasons see the highest participation.

4. Gender-Specific Insights:

   • Men are faster than women across all distances.
   • Women exhibit more consistent speed in races.

---

Visualizations and Results

Suggested Visualizations:

1. Participation by Gender Across Distances:
   A bar chart showing the number of male and female participants in 50km, 50mi, 100km, and 100mi races.
   

2. Age Distribution of Athletes:
   A histogram showing the distribution of athletes’ ages, highlighting the concentration around 30-50 years.
   

3. Average Speed by Gender and Distance:
   A grouped bar chart comparing average speeds of male and female athletes across distances.
   

4. Seasonal Participation:
   A pie chart displaying the percentage of races held in each season (spring, summer, fall, winter).
   

5. Top 5 Events with Fastest Finish Times:
   A horizontal bar chart listing events with the fastest average completion times.
   

---

Quantitative Highlights

• Total records in dataset: 7,461,195.
• Filtered dataset size: 88,064 (USA events, 2019).
• Fastest 50km time: 2.83 hours.
• Longest event time: 120 hours (Across the Years event).
• Popular event: AZT Oracle Rumble 50km with 159 participants.

---

Next Steps

1. Extend analysis to multi-year trends for other regions.
2. Build predictive models for performance based on historical data.
3. Create dynamic dashboards to track trends in real time.

---